<p>The Hurlbert Index of Sexual Desire (HISD; Apt & Hurlbert, 1992) is a widely used 25-item self-report measure of sexual desire in intimate relationships. Despite broad application across clinical and empirical research, its factor structure has received limited scrutiny, and its length constrains use in longitudinal and large-scale survey contexts. The present study developed a psychometrically rigorous short form using a sequential, three-sample validation design. Three independent datasets collected via Amazon Mechanical Turk (N = 500, 389, and 418 individuals in committed relationships) served as training, testing, and replication samples. Item reduction combined exploratory structural equation modeling and bifactor methods. The resulting nine-item HISD-SF was best represented by a bifactor model with a dominant global sexual desire factor and three specific factors: sex drive, perceived desire adequacy, and partner-specific desire. Model fit was excellent in the training sample and acceptable across testing and replication samples. Overall reliability was high (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">$\omega_{cat} \approx .97$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>ω</mi> <mrow> <mi mathvariant="italic">cat</mi> </mrow> </msub> <mo>≈</mo> <mo>.</mo> <mn>97</mn> </mrow> </math></EquationSource> </InlineEquation>) with the majority of reliable variance attributable to the global factor (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">$\omega_{h - cat} \approx .89 - .91$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>ω</mi> <mrow> <mi>h</mi> <mo>-</mo> <mi>c</mi> <mi>a</mi> <mi>t</mi> </mrow> </msub> <mo>≈</mo> <mo>.</mo> <mn>89</mn> <mo>-</mo> <mo>.</mo> <mn>91</mn> </mrow> </math></EquationSource> </InlineEquation>); specific-factor reliabilities were low to trivial, supporting global score interpretation in these samples. Convergent validity was strong, with nomological network associations closely mirroring those of the full HISD. MIMIC-based differential item functioning analyses revealed neither practically meaningful nor systematically replicated item bias across age, relationship length, or gender. The HISD-SF offers a brief, theoretically and empirically derived, and psychometrically defensible alternative to the full scale, while also offering greater conceptual clarity and practical utility for research and clinical contexts.</p>

The Hurlbert Index of Sexual Desire-Short Form: Psychometric Properties

期刊论文
发表日期：

2026年7月2日

查看全文

Brady C. Eisert,
Jared R. Anderson,
Maria F. Portillo

摘要

The Hurlbert Index of Sexual Desire (HISD; Apt & Hurlbert, 1992) is a widely used 25-item self-report measure of sexual desire in intimate relationships. Despite broad application across clinical and empirical research, its factor structure has received limited scrutiny, and its length constrains use in longitudinal and large-scale survey contexts. The present study developed a psychometrically rigorous short form using a sequential, three-sample validation design. Three independent datasets collected via Amazon Mechanical Turk (N = 500, 389, and 418 individuals in committed relationships) served as training, testing, and replication samples. Item reduction combined exploratory structural equation modeling and bifactor methods. The resulting nine-item HISD-SF was best represented by a bifactor model with a dominant global sexual desire factor and three specific factors: sex drive, perceived desire adequacy, and partner-specific desire. Model fit was excellent in the training sample and acceptable across testing and replication samples. Overall reliability was high ( $\omega_{cat} \approx .97$ $ω_{cat} \approx . 97$ ) with the majority of reliable variance attributable to the global factor ( $\omega_{h - cat} \approx .89 - .91$ $ω_{h - c a t} \approx . 89 - . 91$ ); specific-factor reliabilities were low to trivial, supporting global score interpretation in these samples. Convergent validity was strong, with nomological network associations closely mirroring those of the full HISD. MIMIC-based differential item functioning analyses revealed neither practically meaningful nor systematically replicated item bias across age, relationship length, or gender. The HISD-SF offers a brief, theoretically and empirically derived, and psychometrically defensible alternative to the full scale, while also offering greater conceptual clarity and practical utility for research and clinical contexts.